Keywords
Abstract
This study presents a novel brain-computer interface system for the rehabilitation of spinal cord injury-induced upper limb impairment using EEG-fMRI fusion-based motor intention decoding and synchronized functional electrical stimulation and exoskeleton. Twelve chronic cervical SCI patients received 60 sessions across 12 weeks. The multimodal system achieved greater decoding accuracy (87.6±4.2%) compared to single-modality systems. There were significant gains in upper limb function, as evidenced by the rise in GRASSP scores from 19.8±6.4 to 29.5±7.2 (p<0.001), an improvement of 48.9%. Neurophysiological testing revealed prominent cortical reorganization, with enhanced activation of the primary motor cortex (27.4±8.3%), which was positively correlated with functional gain (r=0.76, p<0.001). This treatment surpasses the limitations of conventional rehabilitation approaches in achieving a highly accurate temporal relationship between neural intention and action, thus potentially inducing neuroplasticity through an enduring sensorimotor association. The findings suggest that multimodal BCI systems integrated with other techniques can be an effective choice for upper limb restitution in spinal cord injury.
References
[1] A. Bhatia, K. Manda, A. Potbhare, et al., "Brain-Computer-Spinal Interface Restores Upper Limb Function After Spinal Cord Injury," Nature Communications, vol. 12, no. 1, p. 2979, May 2021, doi: 10.1038/s41467-021-23026-x.
[2] C. Li, Y. Xu, T. Feng, et al., "Fusion of EEG and EMG signals for detecting pre-movement intention of sitting and standing in healthy individuals and patients with spinal cord injury," Frontiers in Neuroscience, vol. 19, p. 1532099, Jan. 2025, doi: 10.3389/fnins.2025.1532099.
[3] E. Sayilgan, "Classifying EEG data from spinal cord injured patients using manifold learning methods for brain-computer interface-based rehabilitation," Neural Computing and Applications, Mar. 2025, doi: 10.1007/s00521-025-11201-w.
[4] F. O. Barroso, N. Yozbatiran, C. Koh, et al., "Brain-Computer interface control of stepping from invasive electrocorticography upper-limb motor imagery in a patient with quadriplegia," Frontiers in Human Neuroscience, vol. 16, p. 1077416, Dec. 2022, doi: 10.3389/fnhum.2022.1077416.
[5] A. De Miguel-Rubio, I. Gallego-Aguayo, M. D. De Miguel-Rubio, et al., "Effectiveness of the Combined Use of a Brain-Machine Interface System and Virtual Reality as a Therapeutic Approach in Patients with Spinal Cord Injury: A Systematic Review," Healthcare, vol. 11, no. 24, p. 3189, Dec. 2023, doi: 10.3390/healthcare11243189.
[6] S. K. Khan, Y. Zhu, L. Lu, et al., "Invasive Brain Computer Interface for Motor Restoration in Spinal Cord Injury: A Systematic Review," World Neurosurgery, vol. 188, pp. 208-219, Apr. 2024, doi: 10.1016/j.wneu.2023.12.097.
[7] A. Eizad, H. Ali, S. Tahir, et al., "A M/EEG-fMRI Fusion Primer: Resolving Human Brain Responses in Space and Time," Neuron, vol. 107, no. 6, pp. 1068-1086, Sep. 2020, doi: 10.1016/j.neuron.2020.07.020.
[8] Y. Zhao, L. Gao, J. Wang, et al., "EEG-controlled functional electrical stimulation rehabilitation for chronic stroke: system design and clinical application," Frontiers of Medicine, vol. 15, no. 4, pp. 596-607, Aug. 2021, doi: 10.1007/s11684-020-0794-5.
[9] L. S. Vargas-Valencia, G. Rodriguez-Bermudez, J. M. Rivera-Acosta, et al., "Exploring EEG-based motor imagery decoding: a dual approach using spatial features and spectro-spatial Deep Learning model IFNet," Frontiers in Neuroinformatics, vol. 18, p. 1345425, Feb. 2024, doi: 10.3389/fninf.2024.1345425.
[10] Z. Yang, L. Zhang, C. Bian, et al., "Classification of EEG Signal Using Deep Learning Architectures Based Motor-Imagery for an Upper-Limb Rehabilitation Exoskeleton," SN Computer Science, vol. 6, no. 1, p. 188, Jan. 2025, doi: 10.1007/s42979-025-03743-6.